Sony reveals 1st NXCAM camcorder: HXR-NX5U

I've seen some very good footage coming out of the Z5 so equal to or better than that is good enough for me.
When factoring in all of the other things that the NX5 has on top of great image quality [based on =/> the Z5] it seem like a winner to me.
 
It's pretty clear (from the chart images in that article - which were recorded using the codec) that the NX5U can record images that can resolve way more than 810 lines of image detail across the frame width. It also seems clear (from the charts) that it can resolve at least a little bit beyond 1440 lines of true image detail pretty cleanly (again, across the frame width - which is not the same as lines per picture height). Aliasing does kick in more with diagonals than with horizontal or vertical lines. Especially in portions of images where there is considerable detail that is significantly composed of lines running vertically (in essence), it does seem pretty clear that recording 1920x1080 should offer some benefit, over recording 1440x1080.
 
I'm thinking I'll test the HMC40's footage (when I get the cam next week) and see how it cuts with the HMC150. If the higher resolution is obvious, then I'll sell the 150 and get the NX5 as my A cam.
 
After due consideration, I cancelled my order with B&H for the Sony NX5 and ordered a Panasonic HMC150. I wanted a 20x zoom and the 128GB ssd unit was also highly desireable along with Vegas Pro 9 dialed in for the NXCAm, and the content management software that comes with the camera. But as an event photographer I am put off by the CMOS imager in the SONY NXCAM representation of still photograhers flash. Sadly, a deal breaker for me.
Rick
 
For some folks CMOS is simply unacceptable (for whatever reasons), and the HMC150 more appropriate (having CCD imaging chips). For those of us who value recording a higher level image detail when shooting the 1080 line formats (as well as a number of other features, like dual media recording), yet are not concerned with rolling shutter CMOS issues, the NX5U would seem a very nice alternative. I'm eager to read the comments from early adopters, when the NX5U starts getting out there in the hands of typical users shooting real-world video.
 
All this flash stuff has me REALLY wanting to test how much of a problem it is, and how the Sony Clipbrowser software deals with the problem...
 
Personally, I don't see what the big deal is about camera flashes and CMOS camcorders. When somebody is shooting stills using a light flash, as you are shooting video, and the flash goes off, it is quite noticeable in the video recording. That's true if you are shooting with a camcorder that uses either CCD or CMOS imaging chips - you get unavoidable overexposure in frames. The only difference is, with a camcorder that uses CCD imaging chips you get entire frames overexposed, and with a camcorder that uses CMOS imaging chips you get a banded portion of frames overexposed. Big whoop. Played back at full speed, to the typical viewer, the perception is about the same - they simply perceive that a light flash went off (and usually, far more often than not, don't even give it a fleeting second thought)!

Videographers who just get highly upset at banded overexposure in a frame, and much prefer frames that are thoroughly overexposed 100% from top to bottom, should definitely purchase camcorders made with CCD imaging chips (and thus avoid any need to see a therapist in order to deal with the resulting emotional distress). For the rest of us, no worries.
 
Sony EX1 with a flash going off can exhibit a loss of up to 7 or so frames. Unacceptable. I'm sure this new cam will be no different. It wouldn't sway me anyway even if it were ccd. I've got Panasonic in my weins!
 
shaun1970 said:
Sony EX1 with a flash going off can exhibit a loss of up to 7 or so frames. Unacceptable. I'm sure this new cam will be no different. It wouldn't sway me anyway even if it were ccd. I've got Panasonic in my weins!
Click to expand...


Isn't it 1/2 frames? That can be corrected in post pretty easily. Hopefully Sony will have a software update like Panasonic did for the HPX300 that makes the full frame flash.
 
Honest to goodness, I've never really counted frames when watching video on television where flashes, from photographers shooting stills, were going off, nor really gave a second thought to whether or not the blowouts in the affected frames (regardless of how many) were banded or not. I seriously doubt 99% of typical viewers do either, but if it bugs the heck out of you to shoot video with banded blowouts or to many of them or whatever with CMOS, when somebody fires a flash to take a still, by all means shoot with a cam that uses CCD imagers. It's a free country, and fortunately we can make the choice of CCD or CMOS, whichever suits us best as individual shooters.
 
It could negatively impact more than 1 frame, depending on where in the recording cycle the flash occurred, because of the long GOP recording format.

"Negatively impact" implies more than just visible flash banding on a part of the frame; the problem is more insiduous because more bitrate is required to account for the difference between frames (i.e. a partial flash frame vs. non flash), thus overall quality deteriorates for a given bitrate during a flash affected GOP sequence. The earlier on in the cycle, the worse the problem
 
PDR said:
It could negatively impact more than 1 frame, depending on where in the recording cycle the flash occurred, because of the long GOP recording format.

"Negatively impact" implies more than just visible flash banding on a part of the frame; the problem is more insiduous because more bitrate is required to account for the difference between frames (i.e. a partial flash frame vs. non flash), thus overall quality deteriorates for a given bitrate during a flash affected GOP sequence. The earlier on in the cycle, the worse the problem
Click to expand...


Oh ya... I hadn't thought about the long gop.

Still should be fairly easy to fix in post if it were that big of an issue.
 
ATL Media Group said:
Oh ya... I hadn't thought about the long gop.

Still should be fairly easy to fix in post if it were that big of an issue.
Click to expand...


Well it depends on what you are shooting and many other variables such as complexity of the content etc..

Sure you can partially fix the visible banding, but you can't recover the lost details from taxed compression during that flash affected GOP.

In AVCHD type compression this usually manifests as deblocking/smearing. GH1 users affectionately know this as "mud". When you hit a non flash affected GOP sequence, details return, so you can get dips and bumps in image quality. Very annoying. You don't have to count frames or pixels to see this; even "normal" folk can see this when it occurs.

Does this mean you will always encounter this? Definitely not , and variable severity. It depends on many factors, shooting conditions etc..

Mind you, this AVCHD implementation should be pretty decent, but I haven't seen any native samples yet
 
That's not a CMOS vs CCD issue. It's an interframe compression issue. Actually, if a flash of light results in an entire frame being overexposed, that is more taxing on a codec (that employs interframe compression) than if the overexposure is limited to a band within the frame.
 
Here's a video [I did not shoot this and this is from the Z5 not the NX5] that has some flashes in the first few seconds and then again around the 2 minute mark.
http://www.vimeo.com/6967820


For me this would not be a deal breaker but if it is too much for other people to work with then I can respect that.
Different people have different needs and different wants so we have different cameras.
 
Hidef1080 said:
Here's a video [I did not shoot this and this is from the Z5 not the NX5] that has some flashes in the first few seconds and then again around the 2 minute mark.
http://www.vimeo.com/6967820


For me this would not be a deal breaker but if it is too much for other people to work with then I can respect that.
Different people have different needs and different wants so we have different cameras.
Click to expand...


Actually that video solidifyed to me that it's not a big deal. Most of the flashes were full screen, not bands.
 
ATL Media Group said:
Actually that video solidifyed to me that it's not a big deal. Most of the flashes were full screen, not bands.
Click to expand...


I noticed that too.
And while the effects or weakness of rolling shutter are real I'm coming to the conclusion that the CMOS thing is being blown out of proportion in many cases.


Again, people know what the need and want [Like I want more resolution than what I'm getting now] but for me this flash banding would not be an issue nor would “whip pans” because it's just something that I've never done.
 
You must log in or register to reply here.
Back
Top